INFECTION AND IMMUNITY, Jan. 2004, p. 468–477 0019-9567/04/$08.00⫹0 DOI: 10.1128/IAI.72.1.468–477.2004 Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Vol. 72, No. 1
Role of Neutrophils in Murine Salmonellosis Ce´dric Cheminay, Dipshikha Chakravortty, and Michael Hensel* Institut fu ¨r Klinische Mikrobiologie, Immunologie und Hygiene, FAU Erlangen-Nu ¨rnberg, D-91054 Erlangen, Germany Received 9 June 2003/Returned for modification 11 August 2003/Accepted 17 September 2003
Gastrointestinal infections with Salmonella enterica serovars have different clinical outcomes that range from localized inflammation to a life-threatening systemic disease in the case of typhoid fever. Using a mouse model of systemic salmonellosis, we investigated the contribution of neutrophils to the innate immune defense against Salmonella after oral infection. Neutrophil infiltration was dependent on the bacterial burden in various infected organs (Peyer’s patches, mesenteric lymph nodes, spleen, and liver). However, the massive infiltration of neutrophils did not allow clearance of an infection with wild-type Salmonella, presumably due to protection of intracellular Salmonella against neutrophil activities. A Salmonella mutant strain deficient in Salmonella pathogenicity island 2 (SPI2) was able to infect systemic sites, but its replication was highly restricted and it did not cause detectable attraction of neutrophils. Neutrophil depletion by antibody treatment of mice did not restore the virulence of SPI2 or auxotrophic mutant strains, supporting the hypothesis that attenuation of the strains is not due to greater susceptibility to neutrophil killing. Our observations reveal that neutrophils have completely different roles during systemic salmonellosis and localized gastrointestinal infections. In the latter conditions, rapid neutrophil attraction efficiently prevents the spread of the pathogen, whereas the neutrophil influx is delayed during systemic infections and cannot protect against lethal bacteremia. infection. Mutant strains deficient in SPI1 function are moderately attenuated in systemic virulence after oral infection of mice but not after intraperitoneal or intravenous infection of mice. In contrast, SPI2-deficient mutant strains are highly attenuated after oral infection, as well as after intraperitoneal infection. Both SPI1 and SPI2 encode type III secretion systems that translocate sets of bacterial virulence proteins into eukaryotic host cells. SPI2 functions are required to influence the intracellular fate of S. enterica serovar Typhimurium, and a role for protection of intracellular Salmonella against antimicrobial activities of the host cell involving reactive oxygen intermediates (29) and reactive nitrogen intermediates (4) has been observed. Thus, SPI2 functions appear to interfere with the innate immune defense against Salmonella. The function of spv genes located on the Salmonella virulence plasmid (SPV) is less well understood. Mutant strains deficient in spv genes exhibit reduced replication in the organs of infected mice (13). Localized intestinal infections by S. enterica serovars lead to a strong inflammatory response that is accompanied by a massive influx of neutrophils to the site of infection (see references 10 and 30 for reviews). Cell culture models (21) and the bovine model of Salmonella-induced gastroenteritis (19) indicate that epithelial cells can respond to Salmonella infection with the release of proinflammatory cytokines. Neutrophils attracted by these signals can migrate through the epithelium and inactivate the pathogen within the intestinal lumen. In the murine model of systemic salmonellosis, intestinal inflammation is not observed, and bacteria rapidly infect Peyer’s patches (PP) via M cells. The subsequent pathogenesis is characterized by uptake of Salmonella by phagocytic cells, an intracellular lifestyle, and ultimately systemic spread. Although the roles of several virulence factors in the interaction with epithelial cells and macrophages have been analyzed in detail, there is only limited information concerning the contribution of neutrophils to immunity against Salmonella
The innate immune system has a crucial function in protection of naive organisms against colonization by pathogenic microbes. This function has special relevance for defense at the intestinal mucosa, where only a single layer of cells separates sterile host tissue from the rich microbial population of the gut and from food-borne pathogens. Although several lines of innate immune defense are present at this border, various bacterial pathogens have evolved strategies to inactivate or escape these defenses (reviewed in reference 18). Salmonella enterica is an important gastrointestinal pathogen that causes diseases ranging from gastroenteritis accompanied by local inflammation and diarrhea to typhoid fever, a live-threatening systemic bacteremia caused by S. enterica serovar Typhi. The latter condition can be studied by using the infection model consisting of S. enterica serovar Typhimurium and susceptible (Nramp⫺/⫺) mice, while infection of humans with this pathogen results in a self-limiting gastroenteritis. Successful colonization of murine hosts by S. enterica serovar Typhimurium depends on the functions of a large number of virulence factors. In addition to metabolic flexibility that allows adaptation to nutritionally limited sites, serum resistance, and the ability to repair damage inflicted by host defense mechanisms, S. enterica serovar Typhimurium has evolved specific and complex virulence traits (reviewed in reference 12). Genes required for invasion of nonphagocytic cells by S. enterica serovar Typhimurium are clustered in a chromosomal region termed Salmonella pathogenicity island 1 (SPI1). A second locus, SPI2, harbors genes that are required for intracellular replication of Salmonella and the ability to cause a systemic
* Corresponding author. Mailing address: Institut fu ¨r Klinische Mikrobiologie, Immunologie und Hygiene, FAU Erlangen-Nu ¨rnberg, Wasserturmstr. 3-5, D-91054 Erlangen, Germany. Phone: 49 (0)9131 8523640. Fax: 49 (0)9131 8522531. E-mail:
[email protected] -erlangen.de. 468
NEUTROPHILS AND SALMONELLA
VOL. 72, 2004 TABLE 1. S. enterica serovar Typhimurium strains used in this study Strain
Relevant characteristics
Source or reference
NCTC 12023 HH104 P5D10 P7B12 P10H1
Wild type sseC::aphT, Kanr spvA::mTn5, Kanr sipC::mTn5, Kanr purD::mTn5, Kanr
NCTCa 16 15 9 15
a
NCTC, National Collection of Type Cultures, Colindale, United Kingdom.
infections. Furthermore, the role of virulence factors during the interaction of Salmonella and neutrophils is not well understood. In general, neutrophils are not considered effective in the control of intracellular pathogens. However, the role of neutrophils in murine salmonellosis is controversial, as recent studies reported that neutrophils contribute to the control of infection (7, 27) but that neutrophils cannot prevent colonization of systemic sites (14). As neutrophils are an important part of innate immunity, we set out to analyze the contribution of neutrophils to protection against systemic murine salmonellosis after oral uptake of the pathogen. We investigated the cellular compositions of various infected lymphoid organs and the role of Salmonella virulence factors SPI1, SPI2, and SPV in the cellular response to infection. Special emphasis was put on a mutant strain deficient in SPI2, since the immune response to this highly attenuated strain is not fully understood. The data presented here provide a comprehensive view of the effect of oral Salmonella infection on neutrophil recruitment. MATERIALS AND METHODS Bacterial strains and growth conditions. S. enterica serovar Typhimurium strains used in this study are listed in Table 1. The bacteria were grown overnight in Luria broth (LB) at 37°C. If required for selection of strains, kanamycin and nalidixic acid were added at concentrations of 50 and 100 g/ml, respectively. For enumeration of CFU, suspensions were plated on Muller-Hinton plates or LB agar plates containing appropriate antibiotics. Infection model. Female C57BL/6 mice were purchased from Charles River Breeders (Sulzfeld, Germany). All mice were kept at the animal unit of the Institut fu ¨r Klinische Mikrobiologie, Immunologie, und Hygiene (Erlangen, Germany) under specific-pathogen-free conditions. The mice used for the experiments were between 8 and 12 weeks old. All experiments were conducted in accordance with the ethical guidelines of the University of Erlangen-Nu ¨rnberg. For infection, bacterial cultures were harvested and washed in phosphatebuffered saline (PBS) (Biochrom), and the concentration was adjusted to the desired value in 300 l of PBS. The mice were infected by gavage with 5 ⫻ 107, 5 ⫻ 109, or 1011 CFU. For intraperitoneal infection, an inoculum containing 5 ⫻ 105 CFU was used. The actual number of viable bacteria in the inoculum was always verified by plating serial dilutions onto LB agar plates in the absence and presence of the appropriate antibiotic. CI model. As previously described (16), mice were infected with a mixture containing 5 ⫻ 109 CFU of the S. enterica serovar Typhimurium wild type (Salmonella WT) and 5 ⫻ 109 CFU of a mutant strain. On days 1 and 3 after infection, mice were sacrificed, and various organs were removed and homogenized in PBS. The bacterial loads of the organs were quantified by plating serial dilutions of the homogenates onto agar plates containing nalidixic acid and agar plates containing kanamycin for enumeration of the wild-type strain CFU and the mutant strain CFU, respectively. The competitive index (CI) was the ratio of the mutant strain to the wild-type bacteria recovered from organs (output) divided by the ratio of the mutant strain to the wild-type strain in the inoculum (input). Neutropenic mouse model. In this study, we used a mouse model in which neutrophil depletion is induced by application of RB6-8C5 monoclonal antibod-
469
ies. Briefly, monoclonal antibodies were prepared from supernatants of the RB6-8C5 hybridoma clone by purification on a HiTrap rProtein A FF column (Amersham Pharmacia). The eluate containing antibody was dialyzed against PBS and sterilized by passage through 0.22-m-pore-size membrane filters. The protein concentration was determined by the BCA protein assay (Pierce); the concentration was adjusted to 1.5 mg/ml with sterile PBS, and the preparation was stored at 4°C. One day before infection, 300 g of RB6-8C5 or PBS (control) was injected into mice. As previously reported (27), this treatment resulted in neutrophil depletion in normal and infected mice for at least 3 days. Neither circulating nor resident populations of macrophages and dendritic cells (DC) were affected by this treatment. Analysis of neutrophil influx by FACS analysis. At different times after infection, various organs were removed; one-half of each organ was used for verification of the infection course, and the remainder was processed for fluorescenceactivated cell sorting (FACS) analysis. In case of the PP and mesenteric lymph nodes (MLN), cell suspensions were prepared by cutting tissues in pieces and then incubating the pieces in RPMI containing 1 mg of collagenase D per ml at 37°C. The suspensions were passed through cell strainers (pore size, 100 m; Falcon) in order to remove debris. For preparation of spleen cell suspensions, pieces of tissue were processed directly by gentle passage through cell strainers. Subsequently, the cells were washed and fixed with 0.4% formaldehyde in PBS. After blocking with Fc block (Becton Dickinson) in PBS containing 10% fetal calf serum and 1% bovine serum albumin, the cells were stained with antiCD11c–phycoerythrin (BD), anti-CD11b–fluorescein isothiocyanate (BD), or anti-Gr1–biotin (BD) as the primary antibody conjugate. Biotinylated antibodies were detected by incubation with streptavidin-allophycocyanin (BD). About 105 cells were analyzed by flow cytometry with a FACSCalibur. Analysis of neutrophil influx by confocal microscopy. The half of the organs remaining after CFU quantification were embedded in Tissue Tek (Dia Tec) and frozen in liquid nitrogen. Cryotome sections were fixed with acetone-methanol, blocked, and stained for Salmonella, neutrophils, and DC, respectively, by using the following combinations: rabbit anti-Salmonella antibody (Difco) and fluorescein-conjugated donkey anti-rabbit antibody (Dianova); rat anti-Gr1 antibody (BD) and rhodamine red X-conjugated donkey anti-rat antibody (Dianova); and Armenian hamster anti-CD11c antibody (BD) and Cy5-conjugated goat antiArmenian hamster antibody (Dianova). Labeling was performed overnight at 4°C for the primary antibodies and for 1 h at room temperature for the secondary antibodies. The sections were mounted in Mowiol/DABCO and analyzed by confocal laser scanning microcopy (Leica). Statistical analysis. The means for the different groups were compared by one-way analysis of variance (Prism 3). In most cases the significance of the difference between a group and the control group was tested with Dunnett’s multiple-comparison t test for unpaired values; the exception was the analysis of neutrophil infiltration in the intraperitoneal infection model, in which the values for the mutant groups were compared to the mean obtained for the wild-type strain. Statistical significance was defined as a P value of ⬍0.05.
RESULTS Neutrophil influx after oral infection with Salmonella. The dynamics in various cell populations were quantified after oral infection of mice with Salmonella WT and a mutant strain deficient in SPI2. In accord with previous data (3, 6, 26), oral infection of C57BL/6 mice with Salmonella WT resulted in rapid bacterial replication in PP, MLN, livers, and spleens (Fig. 1). Mice infected with Salmonella WT were moribund on day 5 or 6 after oral infection. Similar progressions of disease were observed after infection with 5 ⫻ 109 CFU and after infection with 1011 CFU. Oral infection with the SPI2-deficient strain also resulted in the presence of viable bacteria in various organs; however, the number of bacteria remained constant for as long as 20 days after infection. Organs from mice in these infection experiments were analyzed to determine the compositions of various cell populations as described in Materials and Methods. A significant change in the neutrophil population was detected 3 days after infection with Salmonella WT in PP, MLN, and spleens, suggesting that there was an influx of neutrophils after infection of these or-
470
CHEMINAY ET AL.
INFECT. IMMUN.
FIG. 1. Course of infection after oral application of Salmonella WT or an SPI2-deficient mutant strain. At various times after oral infection of mice with Salmonella WT and a mutant strain deficient in SPI2 (SPI2), the mean bacterial loads of various organs were determined. The kinetics of bacterial replication are shown for PP (A), MLN (B), spleens (C), and livers (D).
gans (Fig. 2). In contrast, no differences were detected in the populations of DC (marker CD11c⫹) or macrophages (markers CD11b⫹, Gr1low, and CD11c⫺) (data not shown). After infection with the SPI2-deficient strain, no increase in the neutrophil population was observed in PP, MLN, or spleens on day 1, 3, or 7 after infection (Fig. 2). Under these conditions, the populations of DC or macrophages also remained unchanged (data not shown). For a more detailed characterization of the Gr1-positive population, the expression of CD11b was analyzed. The majority of the Gr1-positive population expressed CD11b (Fig. 2B), indicating that most of the cellular influx consisted of activated neutrophils. These observations were confirmed by analyses of histological sections of MLN by laser scanning confocal microscopy (Fig. 3). On day 1 of infection and prior to neutrophil influx, about 16% of the Salmonella WT bacteria colocalized with Gr1-positive cells that probably represented the resident population. On days 2 and 3 after infection, neutrophil influx occurred in the interfollicular region, where replication of Salmonella WT resulted in the formation of clusters of bacteria. In histological sections from mice infected with the SPI2-deficient
mutant strain, we never observed any alteration in the neutrophil population (data not shown). Neutrophil influx after infection with mutant strains deficient in SPI1 or SPV. To investigate if the absence of neutrophil influx is a specific feature of infection by an SPI2-deficient mutant strain, we performed infection experiments with isogenic strains deficient in invG of SPI1 or spvA of SPV and analyzed the dynamics of cell populations under same infection conditions. Although the organ load of an SPI1-deficent strain in the spleen was lower on day 3 after infection, infected animals also developed a fatal systemic salmonellosis with a delay of 2 or 3 days compared to mice infected by Salmonella WT. The SPV-deficient mutant strain exhibited a highly reduced rate of replication in spleens; however, the rate of bacterial replication was still higher than the rate for the SPI2deficient strain (Fig. 4A). The results for the Gr1high population in spleens of infected mice (Fig. 4B) indicate that infection with an SPI1-deficient mutant strain induced the same neutrophil influx on day 3 after infection as Salmonella WT induced. Similar results were obtained for the Gr1high population in PP and MLN (data not shown). Infection with the SPV-deficient mutant strain did not
VOL. 72, 2004
NEUTROPHILS AND SALMONELLA
471
FIG. 2. Neutrophil influx after a single oral dose of Salmonella WT or a mutant strain deficient in SPI2. The data for infection studies performed with a dose of 5 ⫻ 109 CFU represent the data from four different experiments. Experiments with control, Salmonella WT, and SPI2-deficient (SPI2) groups were always performed in parallel, and each symbol indicates the data for one mouse. For panel A, the means (bars) of the relative neutrophil population levels were analyzed in PP, MLN, and spleens. For panel B, the CD11b⫹ neutrophil subpopulation levels were analyzed. Figure 1 shows the organ loads in these experiments after infection with an dose of 5 ⫻ 109 CFU. One asterisk, P ⬍ 0.05; two asterisks, P ⬍ 0.01.
induce significant neutrophil influx on day 3 after infection; however, on day 7 after infection the neutrophil population in the spleen was comparable to the population observed on day 3 after infection with Salmonella WT or the SPI1-deficient mutant strain. This delayed influx of neutrophils was correlated with the reduced rate of replication of the SPV-deficient mutant strain in the spleen (Fig. 4) and other infected organs (data not shown). In order to further analyze if neutrophil attraction was dependent on the phenotypes of the Salmonella strains used in this study, we used an infection model that did not result in colonization of host organs by the bacteria. Salmonella WT or mutant strains deficient in SPI1, SPI2, or SPV were injected into the peritoneal cavity, and 4 h after infection the influx of neutrophils was quantified by flow cytometry (Fig. 5). During this time, the numbers of bacteria remained similar for Salmonella WT and various mutant strains. Infection with Salmonella WT or various mutant strains induced neutrophil influx to similar extents. Based on these observations, we concluded that Salmonella WT and the mutant strains with defects in SPI1, SPI2, or SPV used in this study are equally capable of inducing an inflammatory response and neutrophil migration. Neutrophil influx is dependent on the bacterial burden in infected organs. The observations described above indicate that the neutrophil influx could mainly be a function of the dose of bacteria in an infected organ rather than a function of the virulence factors. In order to establish this possible corre-
lation in the oral infection model, we performed infection studies with different doses of Salmonella WT and mutant strains in order to compensate for the different rates of replication of the strains. Groups of mice were infected orally with an inoculum of 1011 CFU of the SPI2-deficient mutant strain or with 5 ⫻ 107 or 5 ⫻ 109 CFU of Salmonella WT. On day 3 after infection, the organ load was determined (Fig. 6A and C) and the neutrophil influx was analyzed (Fig. 6B and D). In MLN, similar organ loads (about 5 ⫻ 102 CFU per organ) were detected after infection with 1011 CFU of the SPI2-deficient mutant strain and after infection with 5 ⫻ 107 CFU of Salmonella WT. At these organ loads, no neutrophil influx was detected for either strain. Infection with 5 ⫻ 109 CFU of Salmonella WT resulted in a higher bacterial burden (104 CFU per organ), and a significant neutrophil influx was detected. Analyses of spleens of the same infected animals indicated that infection with 5 ⫻ 107 CFU of Salmonella WT led to a higher bacterial burden than infection with 1011 CFU of the SPI2deficient mutant strain. This indicates that the spleen is more permissive for Salmonella replication than MLN. Again, the extent of the neutrophil population was dependent on the bacterial burden in the spleen (Fig. 6C and D). After intraperitoneal infection, similar bacterial organ burdens of Salmonella WT and the SPI2-deficient strain were observed on day 1 after infection (Fig. 6D), indicating that SPI2-dependent bacterial replication had not started. Infection of the spleen by both strains resulted in similar infiltration of
472
CHEMINAY ET AL.
INFECT. IMMUN.
FIG. 3. Histological analysis of neutrophil influx during infection with a single oral dose of 1011 CFU of Salmonella WT by confocal microscopy. (A, B, and C) Fluorescence observed in the red channel after staining for Gr1. Scale bars ⫽ 200 m. (A) Representative image obtained for MLN of the control mice. (B and C) Representative sections of infected MLN (positive for Salmonella not shown), indicating the neutrophil influx 1 day (B) or 2 days (C) after infection. (D and E) Localization of bacteria (green), neutrophils (red), and DC (blue) for MLN on day 1 after infection (D) (scale bar ⫽ 16 m) and on day 3 after infection (E) (scale bar ⫽ 40 m). Panel D also shows xz and yz sections of an area of colocalization of neutrophils and Salmonella WT. Figure 1 shows the organ loads in these experiments for a dose of 1011 CFU.
neutrophils (Fig. 6F). At later times, mice infected with Salmonella WT developed a fatal systemic salmonellosis, while the bacterial organ burdens of mice infected with the SPI2-deficient strain and the number of infiltrated neutrophils increased only slightly. The observations made in the intraperitoneal infection experiment are in line with the data shown in Fig. 5 and indicate that Salmonella WT and the SPI2-deficient strain induced the same extent of neutrophil infiltration. Effect of neutropenia on the course of systemic infection. In order to investigate the role of neutrophils in controlling the progression of murine salmonellosis after oral infection, neutrophils were depleted prior to infection. Depletion by treatment with monoclonal antibody RB6-8C5 was performed since this model appears to be more reproducible than pharmacologically induced neutropenia (14, 27). The antibody-induced neutropenia was controlled by analysis of the Gr1-positive pop-
ulations in blood and various organs. The effect of neutropenia on bacterial replication was analyzed by using the CI model. The CI model allows sensitive comparisons of the virulence of two bacterial strains in a low number of infected animals (2). Because the conflicting results obtained previously for Salmonella WT and strains deficient in SPV could have been due to the method of neutrophil depletion, as well as to the infection model, we repeated a control with an auxotrophic strain used previously (27). We used a mutant strain of S. enterica serovar Typhimurium deficient in purD that was isogenic with other strains used in this study. In PP and spleens of neutropenic mice, the bacterial burden was higher in infected mice than in control mice (Fig. 7), and similar results were obtained for MLN and livers (data not shown). The replication rate of Salmonella WT in neutropenic mice was increased. However, the CI for WT versus the purD mutant strain in neutropenic
VOL. 72, 2004
NEUTROPHILS AND SALMONELLA
473
FIG. 5. Neutrophil infiltration in the peritoneal cavity in the intraperitoneal infection model. Four hours after injection of PBS or PBS containing 104 CFU of Salmonella WT or a mutant deficient in SPI2 (SPI2), SPV (SPV), or SPI1 (SPI1), the cavity was washed with 3 ml of RPMI. The cells present in the exudates were counted and analyzed for Gr1 staining by FACS. The data are data from two independent experiments.
thioglycolate solution. Both strains were killed to the same extent by murine neutrophils, but they were less susceptible to killing than a laboratory strain of Escherichia coli was (data not shown). These observations indicated that depletion of neutrophils was not sufficient to restore the systemic virulence of highly attenuated mutant strains that were deficient either in metabolic functions (purD) or in a virulence factor important for intracellular survival and replication (SPI2). DISCUSSION FIG. 4. Neutrophil influx and bacterial burden after infection with a single oral dose of Salmonella WT or a mutant strain deficient in SPI1 (SPI1) or SPV (SPV). (A) Bacterial burden in the spleens of infected mice analyzed by plating serial dilutions for determination of the number of CFU. (B) Neutrophil influx in the spleens of mice infected with Salmonella WT or a mutant strain deficient in SPI1 or SPV at a dose of 5 ⫻ 109 CFU. Each symbol indicates the data for one mouse from a representative experiment. One asterisk, P ⬍ 0.05; two asterisks, P ⬍ 0.01.
mice was not increased in PP, MLN, livers, or spleens (Table 2). As previously reported for a systemic infection model, the antibody-induced neutropenia did not restore the virulence of the purD-deficient strain (Fig. 7) in the oral model. Finally, mixed-infection experiments were performed with neutropenic and control mice by using an inoculum consisting of a mixture of Salmonella WT and the SPI2-deficient strain. Three days after oral infection, bacterial counts were determined in various organs of both strains, and the CI were calculated (Table 3). The virulence of the SPI2-deficient strain was not restored in neutropenic mice, and the CI in all organs was higher than the CI in control mice. The susceptibilities of Salmonella WT and the SPI2-deficient strain were analyzed in vitro as previously described (10) by using neutrophils isolated from murine peritoneal exudates 1 day after injection of a
In this study, we analyzed the contribution of neutrophils to the control of systemic murine salmonellosis after oral infection. Using confocal microscopy, we observed that 16% of the bacteria were associated with neutrophils on day 1 after infection. This population was also affected by antibody-mediated neutropenia, which increased the bacterial burden on day 1 after infection. Thus, the resident neutrophil population contributes to the early killing of Salmonella after infection. We observed that in various organs infected by Salmonella an influx of neutrophils was induced and that the dynamics of neutrophil infiltration were similar in various infected organs. The extent of infiltration of neutrophils was largely dependent on the bacterial loads in infected organs. However, the attracted neutrophils were not able to control the massive replication of Salmonella WT or a mutant strain deficient in SPI1. In this infection model, neutropenia was not sufficient to restore the virulence of highly attenuated strains harboring mutations causing auxotrophy or defects in SPI2 functions. We propose that the inability of neutrophils to control Salmonella replication and systemic spread is a consequence of the intracellular lifestyle of Salmonella. In vitro experiments indicated that Salmonella WT and various mutant strains are equally susceptible to killing by neutrophils. However, during host infection, the bacteria rapidly enter safe sites, such as intracellular localizations in epithelial cells, macrophages, and
474
CHEMINAY ET AL.
INFECT. IMMUN.
FIG. 6. Influence of bacterial load on the neutrophil influx after oral infection and intraperitoneal infection. The means of CFU counts in MLN (A) and spleens (C) were compared for three different groups of orally (p.o.) infected mice. One group was infected with 1011 CFU of SPI2-deficient bacteria (SPI2), the second group was infected with 5 ⫻ 107 CFU of Salmonella WT, and a dose of 5 ⫻ 109 CFU of Salmonella WT was used as a positive control for neutrophil infiltration. The neutrophil influx in MLN (B) and spleens (D) was quantified. Furthermore, the bacterial burden of the spleen (E) and the neutrophil infiltration (F) were determined on days 1 and 3 after intraperitoneal (i.p.) infection with 5 ⫻ 105 CFU of Salmonella WT or SPI2-deficient bacteria. Mice infected intraperitoneally with Salmonella WT were moribund between days 2 and 3 after infection and were not analyzed further. Each symbol indicates the data for one mouse from one representative experiment, and the means (bars) were compared to the mean for the uninfected group (two asterisks, P ⬍ 0.01).
DC, and the intracellular population appears to be protected from killing by neutrophils and other immune responses. It was recently observed that in murine DC, the Salmonella population is a static population that can survive for several days (5; C. Cheminay, J. Jantsch, and M. Hensel, unpublished observations). We found that two-thirds of Salmonella WT bacteria were associated with DC in MLN on day 1 after oral infection (data not shown). The intracellular location could also be a requirement for systemic spread. Previous observations indicated that migration to and survival in systemic sites is not dependent on the function of the SPI2 system (26). As survival of Salmonella in DC is independent of the function of various virulence factors (22), this cell population might be involved in the systemic spread of the pathogen. Salmonella strains deficient in SPI2 exhibit highly reduced colonization of PP, and a similar defect was observed in MLN, livers, and spleens. However, under our experimental conditions, mutant strains deficient in SPI2 function were able to infect various organs after oral infection, indicating that the SPI2 function is not essential to initiate systemic spread of Salmonella. In addition, there is a cell population in organs like PP, MLN, livers, and spleens that is permissive for SPI2-dependent bacterial replication. Previous studies have shown that Salmonella is associated with macrophages in livers and
spleens (24, 25). Furthermore, cell culture assays revealed that SPI2-deficient strains exhibit highly reduced intracellular replication in macrophages (6, 16, 23). We suggest that in the various organs colonized by Salmonella during systemic infection, macrophages are the main cellular population that promotes bacterial replication. Defects in bacterial replication in macrophages due to mutations in SPI2 result in highly reduced bacterial burdens in various organs, thereby reducing the virulence in the mouse model. Our data do not support the hypothesis that SPI2 has a specific function in the interference with neutrophil functions. Depletion of neutrophils was not sufficient to restore the defect of an SPI2-deficient mutant strain in replication within the host. This is in contrast to the functions of SPI2 in Salmonella within macrophages, such as protection against antimicrobial activities involving reactive oxygen intermediates, as well as reactive nitrogen intermediates (4, 29). These antimicrobial activities are also generated by neutrophils, but they are mainly directed against extracellular bacteria. We observed that SPI2 functions cannot prevent inactivation of Salmonella by neutrophils in vitro. With a mutant deficient in SPI1-dependent invasion the onset of systemic infection was delayed; a lower number of SPI1-deficient Salmonella cells were able to penetrate the in-
NEUTROPHILS AND SALMONELLA
VOL. 72, 2004
FIG. 7. Effect of neutrophil depletion on organ load with Salmonella WT and a purD mutant stain. Cohorts of neutropenic mice (solid symbols) and control mice (open symbols) were infected with Salmonella WT or an auxotrophic strain deficient in purD. On day 1, 2, or 3 after infection, the mice were sacrificed, and the mean numbers of CFU in the PP (A) or spleens (B) for Salmonella WT (squares) and the purD strain (circles) were determined (n ⫽ 2).
testinal epithelium. However, there is evidence that other mechanisms allow uptake of Salmonella (28). The role of spv genes of the Salmonella virulence plasmid is not fully understood. Mutants deficient in the SPV functions are also able to cause systemic infections after oral infection, but the rate of bacterial replication is greatly reduced. This reduced replica-
TABLE 2. Effects of neutrophil depletion on the course of infection with Salmonella WT and an auxotrophic mutant strain Group
Control Neutropenia a
Mean CI (PurD mutant versus Salmonella WT)a PP
MLN
Spleen
Liver
0.00041 0.00071
0.00182 0.00213
0.00528 0.00219
0.03 0.00087
The CI was determined in various organs 3 days after oral infection (n ⫽ 2).
475
tion might allow sufficient control by the host innate immunity and an adaptive immune response at later stages of infection. The observations on the role of neutrophils in immunity against salmonellosis reported here are in contrast to previous reports. A crucial role of neutrophils has been proposed based on intravenous infection studies with S. enterica serovar Dublin (27). In contrast, our studies based on the natural route of infection indicate that neutrophils have a supportive, but not essential function in controlling systemic murine salmonellosis. In a previous study the workers also investigated the roles of different populations of phagocytes after oral infection of mice with S. enterica serovar Typhimurium (14), and it was observed that the macrophage population had a pivotal role in controlling infection. The critical experimental parameter in the various studies appears to be the route of experimental infection with Salmonella. Intraperitoneal or intravenous infection results in a rapid confrontation between Salmonella and neutrophils present in the circulation. In contrast, infection of host organs after oral infection requires attraction of neutrophils to the site of infection, since the resident cell population is rather small in organs. We did not observe a significant influx of phagocytes on day 1 after oral infection, although Salmonella WT had already colonized various organs at this time. These observations indicate that the inflammatory response to Salmonella infection is not fast enough to prevent further colonization. In contrast, the massive influx of neutrophils to the sites of bacterial infection may contribute to pathogen-induced tissue damage and subsequent multiple organ failure. The route of experimental infection with Salmonella is very important for interpretation of the host response. For example, different results have been obtained for the role of lipopolysaccharide binding protein in immunity to Salmonella depending on the route of infection (oral or intraperitoneal) (11, 31). The role of neutrophils reported here is also in contrast to previous observations based on an intraperitoneal infection model. As salmonelloses are food-borne diseases, we consider the oral route of infection the most suitable route for understanding the functions of various mechanisms of the immune system. Marco et al. (20) reported that neutrophils play an important role in the initial resistance to Listeria monocytogenes after infection by the oral route. It was also observed that Listeria survives and multiplies within neutrophils. In another oral infection model for L. monocytogenes, mice treated with RB68C5 had a 5,000-fold-higher bacterial burden in the walls of the stomach, small intestine, cecum, or large intestine than control mice 3 days after infection (8). Systemic infection resulting from ingestion of L. monocytogenes was also greatly enhanced in neutropenic mice. Both of these studies supported the hypothesis that neutrophils play a critical role in the early defense against oral infection with Listeria, as well as in the control of subsequent systemic infections. In experiments with mycobacteria in which the aerosol infection route was used, CH3/HeJ mice (TLR4 deficient) had a reduced capacity to eliminate mycobacteria from the lungs and to control the spread of infection to the spleen and liver (1). CH3/HeJ mice had 10- to 100-fold-higher bacterial burdens than wild-type mice (CH3/ HeN) and succumbed within 5 to 7 months, whereas most wild-type mice controlled the infection and survived for the duration of the experiment. The lungs of CH3/HEJ mice
476
CHEMINAY ET AL.
INFECT. IMMUN.
TABLE 3. Effects of neutrophil depletion on the course of infection with Salmonella WT and an SPI2-deficient mutant strain Group
Control Neutropenia a
CI (SPI2-deficient mutant versus Salmonella WT) (mean ⫾ SD)a PP
MLN
Spleen
Liver
0.0168 ⫾ 0.0311 0.0014 ⫾ 0.0018
0.0129 ⫾ 0.0091 0.0096 ⫾ 0.0132
0.0178 ⫾ 0.0128 0.0084 ⫾ 0.0166
0.0131 ⫾ 0.0185 0.0043 ⫾ 0.0038
The CI was determined in various organs 3 days after oral infection (n ⫽ 8).
showed chronic pneumonia with increased neutrophil infiltration, reduced macrophage recruitment, and abundant acid-fast bacilli. In this model, high levels of infiltration of neutrophils did not allow control of mycobacterial infections. Data reported here and in various previous studies highlight the different role of neutrophils in systemic and gastrointestinal infections with Salmonella. In pathogen-host models that do not result in systemic infections, infections are characterized by rapid attraction of neutrophils to mucosal sites. The pathogen appears to be unable to reach a safe intracellular location that could protect it from killing by neutrophils. In this situation, the invasion of human epithelial cells triggers the release of proinflammatory cytokines (17, 21). In contrast, systemic infections by Salmonella are characterized by invasion and destruction of M cells of PP. Infection by this route appears to bypass efficient induction of proinflammatory cytokines. We observed induction of proinflammatory cytokines only at later stages of infection when Salmonella had already reached systemic sites and proliferated so that the organ loads were high (unpublished observations). It should be noted that the observations described above resulted from the use of different host models, since a mouse model that resembles localized gastrointestinal infection is currently not available. In conclusion, we investigated the role of neutrophils in the immune response against S. enterica serovar Typhimurium after oral infection of mice. Our observations demonstrate that neutrophils are not able to control systemic salmonellosis and support the hypothesis that Salmonella rapidly enters intracellular locations in host cells that allow replication of the pathogen, as well as systemic spread. Neutrophils make very different contributions during systemic salmonellosis and localized gastrointestinal infections. While rapid neutrophil attraction efficiently prevents the spread of the pathogen in localized infections, the neutrophil influx is delayed during systemic infections and cannot protect against lethal bacteremia. Further work is required to understand how these functions are modulated by chemokines and cytokines. ACKNOWLEDGMENTS We thank Ulrich Schaible (MPIIB, Berlin, Germany) for providing the RB6-8C5 hybridoma cell line and Daniela Ja¨ckel and Claudia Geißler for technical assistance with preparation of monoclonal antibodies. The generous support of our work by Martin Ro ¨llinghoff is gratefully acknowledged. This work was supported in part by DFG grant HE1964/7-1, by a grant from the Johannes und Frieda Marohn-Stiftung at the University of Erlangen-Nu ¨rnberg, and by grant QLK2-1999-00310 from the European Commission. REFERENCES 1. Abel, B., N. Thieblemont, V. J. Quesniaux, N. Brown, J. Mpagi, K. Miyake, F. Bihl, and B. Ryffel. 2002. Toll-like receptor 4 expression is required to control chronic Mycobacterium tuberculosis infection in mice. J. Immunol. 169:3155–3162.
2. Beuzon, C. R., and D. W. Holden. 2001. Use of mixed infections with Salmonella strains to study virulence genes and their interactions in vivo. Microbes Infect. 3:1345–1352. 3. Bumann, D. 2001. In vivo visualization of bacterial colonization, antigen expression, and specific T-cell induction following oral administration of live recombinant Salmonella enterica serovar Typhimurium. Infect. Immun. 69: 4618–4626. 4. Chakravortty, D., I. Hansen-Wester, and M. Hensel. 2002. Salmonella pathogenicity island 2 mediates protection of intracellular Salmonella from reactive nitrogen intermediates. J. Exp. Med. 195:1155–1166. 5. Cheminay, C., M. Schoen, M. Hensel, A. Wandersee-Steinhauser, U. Ritter, H. Ko ¨rner, M. Ro ¨llinghoff, and J. Hein. 2002. Migration of Salmonella typhimurium-harboring bone marrow-derived dendritic cells towards the chemokines CCL19 and CCL21. Microb. Pathog. 32:207–218. 6. Cirillo, D. M., R. H. Valdivia, D. M. Monack, and S. Falkow. 1998. Macrophage-dependent induction of the Salmonella pathogenicity island 2 type III secretion system and its role in intracellular survival. Mol. Microbiol. 30:175–188. 7. Conlan, J. W. 1997. Critical roles of neutrophils in host defense against experimental systemic infections of mice by Listeria monocytogenes, Salmonella typhimurium, and Yersinia enterocolitica. Infect. Immun. 65:630–635. 8. Conlan, J. W. 1997. Neutrophils and tumour necrosis factor-alpha are important for controlling early gastrointestinal stages of experimental murine listeriosis. J. Med. Microbiol. 46:239–250. 9. Deiwick, J., T. Nikolaus, J. E. Shea, C. Gleeson, D. W. Holden, and M. Hensel. 1998. Mutations in Salmonella pathogenicity island 2 (SPI2) genes affecting transcription of SPI1 genes and resistance to antimicrobial agents. J. Bacteriol. 180:4775–4780. 10. Fierer, J. 2001. Polymorphonuclear leukocytes and innate immunity to Salmonella infections in mice. Microbes Infect. 3:1233–1237. 11. Fierer, J., M. A. Swancutt, D. Heumann, and D. Golenbock. 2002. The role of lipopolysaccharide binding protein in resistance to Salmonella infections in mice. J. Immunol. 168:6396–6403. 12. Groisman, E. A., and H. Ochman. 1997. How Salmonella became a pathogen. Trends Microbiol. 5:343–349. 13. Gulig, P. A., and T. J. Doyle. 1993. The Salmonella typhimurium virulence plasmid increases the growth rate of salmonellae in mice. Infect. Immun. 61:504–511. 14. Gulig, P. A., T. J. Doyle, J. A. Hughes, and H. Matsui. 1998. Analysis of host cells associated with the Spv-mediated increased intracellular growth rate of Salmonella typhimurium in mice. Infect. Immun. 66:2471–2485. 15. Hensel, M., J. E. Shea, C. Gleeson, M. D. Jones, E. Dalton, and D. W. Holden. 1995. Simultaneous identification of bacterial virulence genes by negative selection. Science 269:400–403. 16. Hensel, M., J. E. Shea, S. R. Waterman, R. Mundy, T. Nikolaus, G. Banks, A. Vazquez-Torres, C. Gleeson, F. Fang, and D. W. Holden. 1998. Genes encoding putative effector proteins of the type III secretion system of Salmonella pathogenicity island 2 are required for bacterial virulence and proliferation in macrophages. Mol. Microbiol. 30:163–174. 17. Hobbie, S., L. M. Chen, R. J. Davis, and J. E. Galan. 1997. Involvement of mitogen-activated protein kinase pathways in the nuclear responses and cytokine production induced by Salmonella typhimurium in cultured intestinal epithelial cells. J. Immunol. 159:5550–5559. 18. Hornef, M. W., M. J. Wick, M. Rhen, and S. Normark. 2002. Bacterial strategies for overcoming host innate and adaptive immune responses. Nat. Immunol. 3:1033–1040. 19. Jones, M. A., M. W. Wood, P. B. Mullan, P. R. Watson, T. S. Wallis, and E. E. Galyov. 1998. Secreted effector proteins of Salmonella dublin act in concert to induce enteritis. Infect. Immun. 66:5799–5804. 20. Marco, A. J., J. Altimira, N. Prats, S. Lopez, L. Dominguez, M. Domingo, and V. Briones. 1997. Penetration of Listeria monocytogenes in mice infected by the oral route. Microb. Pathog. 23:255–263. 21. McCormick, B. A., S. P. Colgan, C. Delp Archer, S. I. Miller, and J. L. Madara. 1993. Salmonella typhimurium attachment to human intestinal epithelial monolayers: transcellular signalling to subepithelial neutrophils. J. Cell Biol. 123:895–907. 22. Niedergang, F., J. C. Sirard, C. T. Blanc, and J. P. Kraehenbuhl. 2000. Entry and survival of Salmonella typhimurium in dendritic cells and presentation of
NEUTROPHILS AND SALMONELLA
VOL. 72, 2004
23. 24.
25. 26.
27.
recombinant antigens do not require macrophage-specific virulence factors. Proc. Natl. Acad. Sci. 97:14650–14655. Ochman, H., F. C. Soncini, F. Solomon, and E. A. Groisman. 1996. Identification of a pathogenicity island required for Salmonella survival in host cells. Proc. Natl. Acad. Sci. 93:7800–7804. Richter Dahlfors, A., A. M. J. Buchan, and B. B. Finlay. 1997. Murine salmonellosis studied by confocal microscopy: Salmonella typhimurium resides intracellularly inside macrophages and exerts a cytotoxic effect on phagocytes in vivo. J. Exp. Med. 186:569–580. Salcedo, S. P., M. Noursadeghi, J. Cohen, and D. W. Holden. 2001. Intracellular replication of Salmonella typhimurium strains in specific subsets of splenic macrophages in vivo. Cell. Microbiol. 3:587–597. Shea, J. E., C. R. Beuzon, C. Gleeson, R. Mundy, and D. W. Holden. 1999. Influence of the Salmonella typhimurium pathogenicity island 2 type III secretion system on bacterial growth in the mouse. Infect. Immun. 67:213– 219. Vassiloyanakopoulos, A. P., S. Okamoto, and J. Fierer. 1998. The crucial role
Editor: B. B. Finlay
28.
29.
30. 31.
477
of polymorphonuclear leukocytes in resistance to Salmonella dublin infections in genetically susceptible and resistant mice. Proc. Natl. Acad. Sci. 95:7676–7681. Vazquez-Torres, A., J. Jones-Carson, A. J. Baumler, S. Falkow, R. Valdivia, W. Brown, M. Le, R. Berggren, W. T. Parks, and F. C. Fang. 1999. Extraintestinal dissemination of Salmonella by CD18-expressing phagocytes. Nature 401:804–808. Vazquez-Torres, A., Y. Xu, J. Jones-Carson, D. W. Holden, S. M. Lucia, M. C. Dinauer, P. Mastroeni, and F. C. Fang. 2000. Salmonella pathogenicity island 2-dependent evasion of the phagocyte NADPH oxidase. Science 287: 1655–1658. Wallis, T. S., and E. E. Galyov. 2000. Molecular basis of Salmonella-induced enteritidis. Mol. Microbiol. 36:997–1005. Yang, K. K., B. G. Dorner, U. Merkel, B. Ryffel, C. Schu ¨tt, D. Golenbock, M. W. Freeman, and R. S. Jack. 2002. Neutrophil influx in response to a peritoneal infection with Salmonella is delayed in lipopolysaccharide-binding protein- or CD14-deficient mice. J. Immunol. 169:4475–4480.
